Multi-phase flow systems are the interaction between materials in which each has distinct and homogenous physical properties. Each component in the flow that is homogenous in its physical properties and separable is referred to as Phase. In a flow system, phase measurement is important to properly optimize, control, and predict multiphase flow systems. Phase physical properties that are required to optimize of measure a multiphase process are phase distribution, velocity, and mass flow rate. It is desired to obtain such measurements for each phase individually while all phases are interacting in a multiphase flow system. For example, for a multiphase flow system consisting of gas, liquid, and solid particles, measured information of each individual phase is needed to properly control and understand the flow system.
Electrical capacitance sensors use the difference in dielectric constants between phases to measure a capacitance signal. The capacitance signal is a measure of the effective dielectric constant for the mixture of phases. From the measured effective dielectric constant, phase concentrations are inferred. This technique is most effective when the multiphase flow contains only two phases. For three or more phases, it is difficult to identify each phase concentration from a single capacitance measurement.
In the present invention, a method is devised to measure three or more phases in a multiphase flow system by exploiting the Maxwell-Wagner-Sillars (MWS) polarization effect, and using a capacitance sensor. The MWS effect is the change in effective dielectric of a mixture where at least one phase is conductive. Although each phase would have a relatively fixed dielectric constant at frequencies in the quasi-static range, the mixture would have a frequency dependent effective dielectric constant.
Typically, the transition in the frequency domain at which the effective dielectric constant changes depends on the electric properties of the conductive phase that is causing this effect. For mixtures that have multiple phases, multiple frequency transitions may occur where each frequency transition would be related to a phase in the mixture. Measuring capacitance at frequencies higher and lower relative to the transition point provides a measurement that is specifically related to the phase that caused this transition. In other words, a multiphase flow system can be decomposed to its individual phases where each phase is measured distinctly.
This multiphase decomposition approach can be extended to imaging of each phase distribution, distinctly, by using multiple capacitance sensors in formations that provide a tomography measurement. Here, Electrical Capacitance Tomography (ECT) is the reconstruction of material concentrations of dielectric physical properties in the imaging domain by inversion of capacitance data from a capacitance sensor.
Electrical Capacitance Volume Tomography or ECVT is the direct 3D reconstruction of volume concentration or physical properties in the imaging domain utilizing 3D features in the ECVT sensor design. ECVT technology is described in U.S. Pat. No. 8,614,707 to Warsito et al. which is hereby incorporated by reference.
Adaptive Electrical Capacitance Volume Tomography (AECVT) provides higher resolution volume imaging of capacitance sensors based on different levels of activation levels on sensor plate segments. AECVT is described in U.S. Patent Application Publication US2013/0085365 A1 to Marashdeh et al. which is hereby incorporated by reference.
In ECT, ECVT, or AECVT, the capacitance measurement between sensor plates are also related to the effective dielectric content between that plate pair. The phase's decomposition method can be extended to all measurements of ECT, ECVT, or AECVT sensors, thus providing a visual representation of each phase, alone, through image reconstruction.
The relationship between effective dielectric constant and operating frequency is described in several empirical and theoretical approaches that incorporate each phase volume fraction and electrical properties. As each phase electrical properties are known or can be measured, it is the dynamic volume fraction distribution that is sought in multi-phase flow imaging. Here, the instantaneous phase distribution and volume fraction is sought. Among the most notable formulations that relate effective dielectric constant to each phase electric properties and volume fraction are the Wiener, Bruggeman, and Wagner formulations. Those formulations, and all others, take the electric properties of each phase and their volume fraction as inputs to estimate the effective dielectric constant. The available different formulations are related to the nature of the mixing between phases. For example, a system where all phases are homogenously mixed has a different formulation than a system where phases are layered. Nevertheless, all formulations use the same inputs to infer the effective dielectric constant. All effective dielectric constant formulations can be summarized as:εeffective=ƒ((∈1,σ1,φ1),(∈2′,σ2,φ2) . . . (∈n′,σa,φn),ω)Where ƒ is the formulation function, n is the number of phases in the multi-phase flow system, ω is the angular frequency at which capacitance is being measured, and (∈n′,σn,φn) are the complex permittivity, conductivity, and volume fraction of the nth phase, respectively.
Using the effective permittivity formulation, capacitance sensors can be used to image each phase individually in this invention of multi-Phase flow decomposition. As measured capacitance is linearly related to the effective dielectric constant between capacitance plates used to measure capacitance, developed formulation for effective dielectric constants can be used to isolate each phase. Specifically, multiple capacitance measurements are taken at different frequencies. Capacitance acquired using an ECT, ECVT, or AECVT sensor at a single frequency is used to reconstruct an effective permittivity map. As capacitance is acquired at multiple frequencies, a number of permittivity images equal to the number of frequencies will be available. Here, we assume each pixel in the image is a well-mixed region of all phases in the multi-phase flow. The volume fraction of each phase in every pixel is calculated by solving a number of equations equal to the number of phases. Those equations are obtained from formulations that related effective dielectric constant at each frequency.
A data acquisition system operating at multiple frequencies is required for phase decomposition. Capacitances can be measured at different frequencies successively or simultaneously. In the former approach, the data acquisition speed of capacitance values at different frequencies should be higher than flow speed. In the latter, a synchronous demodulator is used to isolate each capacitance value related to each frequency. Using both measuring schemes, the difference between measured capacitances (successive or simultaneous) is used to isolate the change in effective dielectric constant for multi-phase flow decomposition.
In the preferred embodiment, frequency transition points (markers) are first identified. As the effective dielectric constant changes as a function of frequency, points at which there are sharp transitions in the effective dielectric constant can be identified to calculate the effective dielectric constant from a given frequency, volume fraction of each phase, and electrical properties of the mixture. For example, this can be accomplished by using the electrical properties of each single phase alone, and then applying the MWS effect for the mixture based on the volume fraction range for each phase in the mixture. Another method of identifying the frequency transition points involves running a sweep frequency signal of different frequency components and identifying frequency points where the effective dielectric constant of a mixture undergoes a sharp transition. After the frequency points are identified, frequency markers are assigned. A frequency marker is an excitation signal composed of one frequency above and below the identified transition point. For multiple transition points, multiple markers can be used. Each frequency marker measures the capacitance at a frequency above and below a transition point. The difference in measured capacitance at different frequency markers is related to the volume fraction of the phase that introduced this frequency transition phenomenon in the effective dielectric value. As each phase has different electrical properties, the points at which each phase contributes to a sharp transition in the effective dielectric constant of the whole mixture is distinct in the frequency domain, as is the identification of frequency markers for each phase.
The measurements of capacitance difference for each phase marker are used to reconstruct 3D images of phase distribution in the imaging domain. Following this approach, one can obtain more than one phase distribution image, each corresponding to a different phase. The number of different phase distribution images that can be obtained from the phase separation approach depends on the number of frequency transition points and phase markers identified. Each single phase volume image can then be used to reconstruct a global image where all phases are visible.
The need for a phase separation method is eminent in capacitance tomography. As capacitance measurements are related to the effective dielectric constant between sensor plates, a maximum of two phases can be imaged when capacitance measurements at a single frequency are used. Moreover, a number of physical properties can't be inferred directly from the conventional capacitance imaging method. For example, the velocity of each phase in a multiphase flow system varies based on the physical properties of each phase. It is desirable to measure the velocity of each phase to better control and understand the processes being measured. Capacitance based imaging provides a velocity measurement of the mixture, but not of each phase alone. For applications that involve mass flow gauging (i.e. oil logging and transportation in pipelines or solids/liquid mass flow rates in fluidized beds) or those requiring a distinction between velocities of different phases, conventional capacitance sensors fall short. The phase separation method developed here fills the gap by providing a method of imaging and measuring each phase independently, fulfilling a practical need for phase separation of multi-phase flow. Other applications that can utilize this technology are multiphase processes that use catalysts and imaging of the human body. Both of those examples involve phases that are conductive, thus enabling this phase separation approach. However, in cases where different layers are layered rather than mixed, the disclosed approach may be used to identify boundaries between phases. For example, applications that involve the human body where tissues of different properties are layered, the disclosed invention can be used to isolate each layer of similar tissues based on identifying its boundaries.
The present invention provides an innovative ECVT sensor and supporting features for multi-phase flow decomposition based on multi-frequency application. This decomposition utilizes ECT, ECVT, or AECVT sensors with a data acquisition unit that can measure capacitances at different frequencies. In the preferred arrangement, a signal of multiple frequencies is sent to the sensing plate. The receiving plate employs several synchronous demodulations in parallel to measure the current at the receiving plate at each frequency. The phase decomposition method is then employed to extract information and generate volume images of each phase individually. The frequencies that correspond to a specific phase are identified based on electric properties of that phase, and other mixing phases in the multiphase flow system. As each phase has different electrical properties, the points at which each phase contributes to a sharp transition in the effective dielectric constant of the whole mixture is distinct in the frequency domain, as is the identification of frequency markers for each phase.
The present invention also provides a method for identifying boundaries between different phases in the layered structure. For example, the human body is formed from different layers (skin, fat, bone etc.) that are layered from the outside inward. As different layers have different electrical properties (dielectric constant and conductivity), the MWS effect will take place at the interface between said layers. The phase decomposition method can be used to identify boundaries between layers for better imaging. Those boundaries can also be integrated in a global volume image where all phases are viewed simultaneously. The integrative and adaptive data acquisition method is used to activate the ECVT sensor plates with different frequencies for detection and volume identification of rusted steel or broken steel cable strands.
An Adaptive data acquisition system can be extended to a new formation that enables simultaneous capacitance measurement of capacitance at different frequencies, as depicted in FIG. 3. Technical features of the system reported in PCT/US14/24457 can also be applied to the current formation of block diagrams in the new formation reported in this pattern.
In (Ser. No. 14/564,204), a dual frequency approach was devised to image steel rust in tendons. The present invention is different in the following ways: 1) it employs more than two frequencies to detect multiple phases in the flow (in prior technology only a single phase was addressed); 2) it devises a method to establish frequency markers based on knowledge of electrical properties for different phases; 3) a method is also devised to quantify phases based on MWS effect; 4) image reconstruction:
a—Capacitance is measured at all frequency markers for all combinations of plate formations;
b—An image is reconstructed using any available reconstruction (algebraic or optimization) technique for each set of capacitance measurements corresponding to a specific frequency;
c—Each pixel or voxel (volume pixel) corresponding to the same location in all reconstructed images is considered for phase decomposition in that voxel. Here, the effective dielectric constant reconstructed at each image will change based on the frequency used to measure the capacitance and generate the image;d—The difference between effective dielectric constant in voxels located in the same location of each reconstructed image (across all images) is used to identify each phase volume fraction. Here, the effective dielectric constant for a well-mixed multiphase flow system is formulated by relating the effective dielectric constant to the electric properties of phases, the frequency used to measure capacitance, and the volume fraction of each phase. Since all involved parameters are known expect for each phase volume fraction, a number of equations is formed for each voxel location. The number of those equations is equal to the number of frequency points at which measurements were conducted. The set of equations is then solved to calculate the volume fraction of each phase for every voxel (this will allow quantification of the percentage of volume occupied by each phase in the voxel);e—The process in d is repeated for each voxel location across all images reconstructed for capacitance measurements at different frequencies. Here, volume fraction of each phase is calculated at each voxel location, combining the volume fraction of a single phase in all voxel locations provides a volume image of that phase alone;f—This phase decomposition process starts with a set of images reconstructed at different frequencies and ends at the same number of images, but each one corresponding to a single phase only. Individual phase volume fractions for each pixel (voxel) are combined together to form an image of that single phase alone. See steps above.